Datasets:

Modalities:
Text
Formats:
json
Languages:
Norwegian
ArXiv:
Libraries:
Datasets
pandas
License:

Possibly missing citation

#3
by emanlapponi - opened

Hi there!

I noticed this dataset is exactly the same timeframe as https://github.com/ltgoslo/talk-of-norway, was it possibly extracted from that curated dataset? In that case, a citation in the model card would be appropriate.

Besides, the card states that Fremskrittpartiet and SV are the two major parties in Norway, which is not the case.

Nasjonalbiblioteket AI Lab org

Pinging @pere , who created the dataset.

Nasjonalbiblioteket AI Lab org

Thanks for the feedback. This is a small classification dataset made from https://www.nb.no/sprakbanken/ressurskatalog/oai-repo-clarino-uib-no-11509-123/. This should have been stated clearer. I also added a link to https://github.com/ltgoslo/talk-of-norway, though I am not entire sure what the differences are between these two resources.

I also agree that "the two major parties" is not a good formulation. I guess what I tried to write was "two of the major parties", but I ended up rewriting the entire section. Hope it is clearer now.

I did a major update of the entire data card here.

Nasjonalbiblioteket AI Lab org

Another typo in the link here. Fixed now.

Nasjonalbiblioteket AI Lab org

Thanks for the comments on this @emanlapponi . If I missed something about the differences between the LTG-git and what is in Ressurskatalogen, please tell me, and I will update the page.

Thank you @pere ! Wasn't aware of Kummervold et al., who refer to the correct original data set.

So all good! And thanks for the shoutout in the card 🤗

emanlapponi changed discussion status to closed

Sign up or log in to comment